Within the context of research and practice with self-expression techniques, the Thematic Apperception Test (TAT) presents several advantages widely acknowledged. However, in spite of well-developed research with this instrument being seen in other countries, no recent Brazilian studies have been carried out up to date. Such situation arise the need of efforts in order to show the properties of the TAT in the current Brazilian cultural context, so that its strengths and weaknesses can be known. In this sense, the development of norms on the typical performance of a non-clinic population is an important first step toward resuming the Brazilian studies on the TAT, at the same time such norms could be used by applied professionals. Considering that the validation of self-expression techniques is related to classification systems (and not the instruments themselves), the present project adoptsamong the myriad of categorizing systems for TAT stories-the one proposed by Monique Morval, which was developed for students and professionals with no experience with the TAT. Such system adopts assumptions from Henry Murray's Personology, Cognitive Psychology and Ego Psychology, providing both objective and qualitative data on the stories. Due to resemblance of this system with the one currently approved for use by the Evaluation System of Psychological Tests (Sistema de Avaliação dos Testes Psicológicos, SATEPSI) of the Brazilian Federal Council of Psychological, there is a potential for the adoption of such system by Brazilian psychologists. In order to carry out the present project, the now called morvalian system was improved by revising and adding new categories and subcategories, which was based on both Brazilian and international studies regarding. The present project, thus, aimed to develop norms for the TAT (morvalian system) in adults from a randomized, non-clinical sample, which was stratified in terms of age, gender, schooling and socialeconomical levels. The study was carried out in the city of Ribeirão Preto (Brazilian state of São Paulo), with a n = 100 sample (25-44 years of age). Participants were invited to take part of the study during visits to randomly selected houses, located in regions of different socialeconomical levels. Participants were individually assessed, using a semi-structured interview script, a measure of intelligence (the TIG-NV-General Non-Verbal Intelligence Test) and 20 TAT cards, selected according to the original manual of this instrument. Data was compiled with the morvalian system's categories and went through both statistically and qualitative analyzes in search for performance differences related to gender, schooling and socialeconomical levels. The results showed few differences across the groups, which do not support for the elaboration of specific norms. However, results also showed a significant (although small) trend for better performance of groups with better schooling and socialeconomical levels on the variables related to stories' formal organization. Such results are discu...